Title: Requirements on Future Spaceborne SAR Systems for Maritime Surveillance
Authors: VAN WIMERSMA GREIDANUS HERMAN
Citation: Advanced SAR Workshop 2007
Publisher: Canadian Space Agency
Publication Year: 2008
JRC Publication N°: JRC37319
URI: http://www.space.gc.ca/asc/eng/events/2007/asar.asp
http://publications.jrc.ec.europa.eu/repository/handle/JRC37319
Type: Contributions to Conferences
Abstract: This paper discusses requirements coming from (potential) users of satellite-based vessel surveillance on future satellite SAR systems. It is an output of the 24-partner DECLIMS project that was concerned with commercial satellite imagery for maritime vessel detection and classification. In order to derive requirements, 27 scenarios for maritime surveillance are analysed, relating to application areas ranging from fisheries control to maritime security. Functional requirements are derived on minimum vessel size to be detected, instantaneous swath width, the need to obtain vessel attributes (such as type, size, velocity), revisit frequency, lead and delay times, and aspects of data quality, availability and reliability. Based on the experience with existing systems, some possibilities for technical improvements are discussed that may help in addressing the requirements. All scenarios have somewhat different requirements in detail but there are some clear trends and commonalities. It is not possible to state hard numbers (e.g., a scenario or user claiming to need a 200 km swath width would probably still be satisfied with a 170 km swath). In many cases, certain system capabilities can serve a scenario’s basic needs, but improved capabilities would serve the scenario much further. Most scenarios need a wide instantaneous swath, up to 400 km or more. Some of them need this in combination with the capability to detect small vessels, which is a challenge. Minimum vessel sizes of interest vary between 50 m and 3 m. Some of the latter scenarios are not accessible to spaceborne imaging, especially in combination with wide swath or very high revisit times (they would need capability improvements which cannot be expected in the near or mid-term future). Most of the scenarios require that vessel type and size are obtained; something which is often not possible with today’s spaceborne SAR systems. In many cases there is a requirement to estimate ship speed and course, which, again, is not routinely possible today. Requirements for revisit time are for many scenarios of the order of a few days, but a decrease in revisit time to 12 or even 6 hours would greatly benefit many applications. A similar situation exists with regard to the delay time for delivering the results after image acquisition, and the lead time to task the satellite. Many scenarios require a few hours maximum delay and 3 days ahead tasking; but less than 20-30 minutes delay and 12-6 hours tasking lead time would open up many more possibilities. Results delivery refers to the ship detection and classification outcome. Geographical coverage of interest is essentially worldwide (including the short delay times). Performance shortfalls of existing systems have been signalled in DECLIMS, and they directly provide requirements for future systems. These requirements include (in addition to points already mentioned above): improvement of detection probability for smaller vessels and decrease of false alarm rate, also under adverse weather conditions and close to land (which are especially problematic now); to have quantitative knowledge about actual detection probability and false alarm rate (which for SAR implies a requirement for absolute calibration of the image); countering azimuth smearing in SAR images (currently frustrates size estimation); availability of the system when required (at present problematic due to mode conflicts); and more easily accessible meta-data.
JRC Institute:Institute for the Protection and Security of the Citizen

Files in This Item:
There are no files associated with this item.


Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.